Generative AI and Multimodal Learning for Early Gastric Cancer Detection Using Small, Imbalanced, and Heterogeneous Clinical Data

Gastric cancer remains one of the leading causes of cancer-related mortality worldwide, primarily due to late-stage diagnosis. Early detection significantly improves survival rates, yet current diagnostic approaches rely heavily on endoscopic expertise and histopathological analysis, which are resource-intensive and subject to inter-observer variability. Artificial intelligence (AI) has shown promise in medical image analysis, but its application in gastric cancer detection faces critical challenges. Clinical datasets are often small, imbalanced across disease stages, and heterogeneous due to variations in imaging protocols and patient demographics. Most existing AI models are trained on large, balanced datasets and fail to generalize under these real-world constraints. There is an urgent need for robust AI methodologies specifically designed to handle limited, imbalanced, and heterogeneous clinical data to enable reliable early gastric cancer screening across diverse healthcare settings.

This Research Topic aims to advance AI-driven early gastric cancer detection by addressing the fundamental challenges of small sample sizes, class imbalance, and data heterogeneity in clinical environments. While deep learning has achieved remarkable success in medical imaging, its clinical translation remains limited due to over-reliance on large, well-curated datasets that do not reflect real-world conditions. Recent advances in generative AI, including Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), and diffusion models, offer new possibilities for synthetic data generation and minority class augmentation. Simultaneously, multimodal learning frameworks enable the integration of diverse data sources such as endoscopic images, histopathology slides, and electronic health records, potentially improving diagnostic robustness. We seek contributions that leverage these technologies to develop AI systems capable of learning effectively from limited data, adapting across clinical sites, and providing interpretable outputs for clinical decision support. The goal is to bridge the gap between experimental AI models and clinically deployable solutions for early gastric cancer detection.

We welcome original research, reviews, and methodological papers addressing, but not limited to, the following themes:

- Generative AI techniques (GANs, VAEs, diffusion models) for medical image synthesis and class imbalance correction.
- Multimodal fusion methods integrating endoscopic, histopathological, and clinical data.
- Small-sample learning approaches including few-shot, meta-learning, and transfer learning for medical imaging.
- Explainable AI (XAI) methods for clinically interpretable diagnostic models.
- Domain adaptation and generalization techniques across hospitals, populations, and imaging systems.
- Real-world clinical validation and multi-center studies of AI-assisted gastric cancer detection.
- Comparative analyses of generative versus discriminative learning in medical contexts.

We encourage submissions of Original Research, Systematic Reviews, Brief Research Reports, and Perspective articles that emphasize practical, clinically translatable AI solutions for early gastric cancer detection.

Article types and fees

This Research Topic accepts the following article types, unless otherwise specified in the Research Topic description:

• Brief Research Report
• Case Report
• Clinical Trial
• Community Case Study
• Editorial
• FAIR² Data
• General Commentary
• Hypothesis and Theory
• Methods
• ... View all formats

Articles that are accepted for publication by our external editors following rigorous peer review incur a publishing fee charged to Authors, institutions, or funders.

Article types

This Research Topic accepts the following article types, unless otherwise specified in the Research Topic description:

• Brief Research Report
• Case Report
• Clinical Trial
• Community Case Study
• Editorial
• FAIR² Data
• General Commentary
• Hypothesis and Theory
• Methods
• Mini Review
• Opinion
• Original Research
• Perspective
• Review
• Study Protocol
• Systematic Review

Keywords: Gastric Cancer, Generative Artificial Intelligence, Multimodal Learning, Small Sample Learning, Imbalanced Medical Data, Medical Image Analysis, Explainable AI, Clinical Decision Support, Medical Image Synthesis, Data Scarcity, Synthetic Data

Important note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.